INDEX

Explanations

generic words and phrases such as "way to", "are", "any", and "looking for" that appear often in questions and helpful answers

oai_token-act-pair · gemini-2.0-flash

variations of the word "way"

np_token-act-pair-logits · gpt-4o-mini

"way"

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AddTagHelper

-0.68

twimg

-0.65

Personensuche

-0.57

enumii

-0.56

UpInside

-0.54

 للمعارف

-0.52

BASELINE

-0.50

 kaarangay

-0.50

 unknownFields

-0.49

ristor

-0.48

POSITIVE LOGITS

way

2.92

 ways

2.53

Way

2.17

way

2.17

WAY

2.14

Way

2.13

 Ways

2.02

Ways

1.96

 WAYS

1.93

 means

1.91

Activations Density 5.107%

generic words and phrases such as "way to", "are", "any", and "looking for" that appear often in questions and helpful answers

variations of the word "way"

"way"

No Comments

No Known Activations